THE APP : goals and characteristics
Human emotion recognition plays an important role in the interpersonal relationship. Emotions are reflected from speech, hand and gestures of the body and through facial expressions. Hence extracting and understanding of emotion has a high importance of the interaction between human and machine communication.
Emotion plays indeed an important role in human life. Interpersonal human communication includes not only language that is spoken, but also non-verbal cues as hand and body gestures, tone of the voice, which are used to express feeling and give feedback and most importantly through facial expression. Human beings express emotions in day to day interactions. Understanding and knowing how to react to people's expression greatly enriches the interaction. The field of psychology has played an important role in understanding human emotion and developing concepts that may aid these HCI technologies. Ekman and Freisn have been pioneers in this area, helping to identify six basic emotions : anger, fear, disgust, joy, surprise and sadness. These emotions appear to be universal across humanity. Emotion recognition comes under computer vision. Computer vision seeks to generate intelligent and useful descriptions of visual scenes by performing operations on the signals received from video cameras.
Anger, Fear, Disgust, Joy, Surprise and Sadness.
These are the emotions that our app will be able to identify after exercising. With this artificial intelligence, such an application will be helpful for robots interacting with humans, but also in many other fields such as marketing. Eventually, the application will be able to determine the "expressive index of pleasure", a logical consequence of the emotion detection of its users. Hence, this app could be the first one which can evaluate satisfaction among human. A huge progress especially when people know the low reliability of the satisfaction curves.
MACHINE LEARNING PROGRAMMATION
Ml5.js is a technology that enables programmers to include learning abilities to an application.
Logic
MobileNets are a class of convolutional neural network which are trained to recognize images. The more images the AI recieve, the more it learns to recognize them, the more able it is to differentiate them by comparison. By using this neural network, the app will give a percentage of confidence to its assertion.
Process
First, we should help the AI to analyze a set of images showing different facial expressions (create a databases images as reference for the AI). The AI should be programmed to recognize the visible or more subtile characteristics of these emotions such as smiles, smirk, frowning, blank stares and so on.
Then, the users should be able to capture on live any human face whose facial expression need to be analyzed. Tnahks to the real-
Resources
How to code an app with MobileNets functionality ? Click here
To understand more about MobileNets :
A project to facilitate the future relationships between humans, human and robots. A way to integrate psychology and eventually more ethics inside progress.
BY SOUKAYNA KEBIRI